Openlinksoftware MCP Sqlalchemy Server

🚀 MCP Server for ODBC Implemented via SQLAlchemy

A lightweight MCP (Model Context Protocol) server for ODBC, built using FastAPI, pyodbc, and SQLAlchemy. This server is compatible with the Virtuoso database management system (DBMS) and other backend databases that implement SQL Alchemy providers.

🚀 Quick Start

This is a lightweight MCP server for ODBC. It is built with FastAPI, pyodbc, and SQLAlchemy, and is compatible with Virtuoso DBMS and other relevant backend databases.

✨ Features

• Get Schemas: Fetch and list all schema names in the database.
• Get Tables: Retrieve table information for a specific schema or all schemas.
• Describe Tables: Generate a detailed description of the table structure, including:
  • Column names and data types
  • Nullable attributes
  • Primary and foreign keys
• Search Tables: Filter and retrieve tables based on name substrings.
• Execute Stored Procedures: For Virtuoso, execute stored procedures and obtain results.
• Execute Queries:
  • JSONL result format: Optimized for structured responses.
  • Markdown table format: Suitable for reporting and visualization.

📦 Installation

Prerequisites

1. Install uv:

   pip install uv
   

   Or use Homebrew:

   brew install uv
   

2. Check the unixODBC runtime environment:

   1. Check the installation configuration (e.g., the location of the INI file): Run odbcinst -j
   2. List available data source names: Run odbcinst -q -s

3. Set the ODBC data source name: Configure your ODBC data source name (~/.odbc.ini) to match the target database. Example for Virtuoso:

   [odbc]
   Driver = Virtuoso ODBC Driver
   ServerName = localhost:8889
   DatabaseName = db_name
   Username = user_name
   Password = user_password
   

4. Install SQLAlchemy and Python ODBC libraries:

   pip install sqlalchemy pyodbc
   

Steps

1. Clone the repository to a local directory:

   git clone https://github.com/your-username/mcp-server.git
   cd mcp-server
   

2. Update the config.py file to match your database settings.

3. Run the server:

   uvicorn main:app --reload
   

📚 Documentation

Configuration

1. Update environment variables: Create or edit the .env file in the root directory and add the following content:

   DATABASE_URL=odbc://username:password@localhost:8889/db_name
   

2. Configuration example for Claude Desktop users:

   npx @modelcontextprotocol/core --directory . run mcp-server
   

Database Connection URL Format

The following table lists the supported databases and their corresponding ODBC connection string formats:

Tools

• virtuosoSupportAI: Interact with the Virtuoso support assistant/agent, a specific feature of Virtuoso for interacting with large language models (LLMs).

Troubleshooting

For easy troubleshooting:

1. Install the MCP inspector:

   npm install -g @modelcontextprotocol/inspector
   

2. Start the inspector:

   npx @modelcontextprotocol/inspector uv --directory /path/to/mcp-server run mcp-server
   

Access the provided URL to troubleshoot server interaction issues.